28 comments

[ 4.2 ms ] story [ 66.8 ms ] thread
OP here. this is a cursed project lol, but i wanted to see: What happens if you replace the OS scheduler with an LLM?

With Groq speed (Llama 3 @ 800t/s), inference is finally fast enough to be in the system loop.

i built this TUI to monitor my process tree. instead of just showing CPU %, it checks the context (parent process, disk I/O) to decide if a process is compiling code or bloatware. It roasts, throttles, or kills based on that.

Its my experiment in "Intelligent Kernels" how they would be. i used Delta Caching to keep overhead low.

You can't replace the NTOS scheduler. This is more of an automated (?) process manager.
I do wonder how painfully slow the computer would be if you actually did replace the in-kernel scheduler with an LLM...
This is a pretty funny project, you've outsourced the neurotic developers that keep their task manager open and kill off processes they don't like.

I wouldn't call it replacing the scheduler though - more that you've made a scheduler manager.

I resemble that comment!

But seriously, it does really bug me on principle that DropBox should use over half a GB simply because it uses Chromium, even when nothing is visible.

Maestral is a cross platform implementation of the Dropbox client API which I use on low end Linux machines.
For me it's LSP servers taking 2 gigs of RAM. With Antigravity, Google managed to go beyond this, it is totally unusable for me (but other VScode clones work fine, apart from the 2 Go LSP servers).
haha exactly. i realized i spent too much time staring at htop wondering what is this process?, so i decided to automate my own anxiety.

Scheduler Manager is definitely the more accurate term. Im just the middleman between the chaos and the kernel.

Now we need processes to gain awareness of the process manager and integrate an LLM into each process to argue with the process manager why it should let them live.
Assistant to the scheduler manager
Assistant scheduler manager
It really is cursed to be spending hundreds of watts of power in a datacenter somewhere to make a laptop run slightly faster.
oh absolutely. burning a coal plant to decide if i should close discord is peak 2025 energy. strictly speaking, using the local model (Ollama) is 'free' in terms of watts since my laptop is on anyway, but yeah, if the inefficiency is the art, I'm the artist.
Running ollama to compute inference uses energy that wouldn't have been used if you weren't running ollama. There's no free lunch here.
An interesting thought experiment - a fully local, off-grid, off-network LLM device. Solar or wind or what have you. I suppose the Mac Studio route is a good option here, I think Apple make the most energy efficient high-memory options. Back of the napkin indicates it’s possible, just a high up front cost. Interesting to imagine a somewhat catastrophe-resilient LLM device…
Macs would be the most power efficient with faster memory but an AI Max 395+ based system would probably be the most cost efficient right now. A Framework Desktop with 128GB of shared RAM only pulls 400W (and could be underclocked) and is cheaper by enough that you could buy it plus 400W of solar panels and a decently large battery for less than a Mac Studio with 128GB of RAM. Unfortunately the power efficiency win is more expensive than just buying more power generation and storage ability.
An entire datacenter on the other hand, might be appealing to spot things you wouldn't otherwise see in a sea of logs and graphs.
You're underselling this as a process manager, it could also be a productivity tool with some prompt changes; Determine procrastination apps: games, non-professional chat, video streaming and kill it.
Task manager, not scheduler.
Please add Roulette mode where a random process is killed every so often
You did not replace the OS process scheduler with an LLM.
This is the one place that I would want Copilot running. It's giving me ideas :)
If it doesn’t find a process that needs roasting or killing for a while, will it see itself as bloatware and commit suicide?